Objective To evaluate the efficacy and safety of thiazolidinediones for metabolic syndrome.Methods Up through 2007, we searched The Cochrane Library, MEDLINE, EMbase, the China Biological Medicine Database, VIP and CMAC. We also handsearched relevant literature. Randomized controlled trials about usingthiazolidinedioes to treat metabolic syndrome were included. Two reviewers independently extracted the data from the eligible studies and evaluated the quality of the included studies. Meta-analysis was performed for the results ofhomogeneous studies using RevMan 4.2.9 software. Results Ten randomized control trials involving 1,183 patients with metabolic syndrome met the inclusion criteria. Meta-analysis was not carried out because of apparent heterogeneity. Five trials compared rosiglitazone and placebo, which of single study reported CVD events at the end of 9 month follow-up. The results suggested that no significant differences were found between the two groups in occurrence of CVD events (RR=0.50, 95%CI 0.25 to 1.00), such as myocardial infarction and urgent vessel revascularization after coronary stent implantation, in the patients with metabolic syndrome, while rosiglitazone significantly decreased the proportion of metabolic syndrome (RR=4.0, 95%CI 1.63 to 9.82) and HOMA-index (WMD=-0.80, 95%CI -0.90 to -0.70) as compared with placebo. Pioglitazone did not affect TG, significantly decreased HOMA-index (WMD=0.02, 95%CI 0.01 to 0.03), and increased HDL-c (WMD=0.02, 95%CI 0.01 to 0.03), compared with placebo. Pioglitazone plus glimepirde was better than rosiglitazone plus glimepiride in TG and HDL-c improvement, with no significant differences in improving BP, FPG, PPG, HbA1c, and HOMA-index for both treatments. The combination of rosiglitazone with metformin was similar to pioglitazone-metformin combination in improving FPG, PPG, HbA1c and HOMA-index, whereas pioglitazone plus metformin was superior to rosiglitazone plus metformin in improving TG and HDL-c. No differences between rosiglitazone-metformin combination and glimepirde-metformin combination were observed in improving FPG, PPG, and HbA1c, but rosiglitazone plus metformin significantly lowered HOMA-index and SBP/DBP more than glimepirde plus metformin. The results of included trails revealed that rosiglitazone and pioglitazone had no favorable effects on BMI and WC or resulted in weight gain. The adverse drug reactions for thiazolidinediones were mild to moderate, and well tolerated. Conclusion The results suggest that thiazolidinediones produce positive effects on blood glucose level and insulin sensitivity in the absence of favorable obesity effects or resulting in weight gain. Pioglitazone favorably affects HDL-c. Thiazolidinediones show a certain effect on decreasing the proportion of metabolic syndrome, but the therapeutic effect on BP is uncertain. Overall there is insufficient evidence to recommend the use of thiazolidinediones for metabolic syndrome due to low methodological quality, small sample size, and limited number of trials. More high-quality, largescale randomized controlled trials are required.